From a9f5e9a5c6e0b069d95158b85b9cafdd67084cba Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9D=D0=B8=D0=BA=D0=B8=D1=82=D0=B0=20=D0=9F=D0=BE=D1=82?= =?UTF-8?q?=D0=B0=D0=BF=D0=BE=D0=B2?= Date: Sat, 21 Dec 2024 04:56:36 +0400 Subject: [PATCH] Lab4 done --- lab_4/.gitignore | 1 + lab_4/lab4.ipynb | 5160 ++++++++++++++++++++++++++++++++++++++++ lab_4/requirements.txt | Bin 0 -> 2706 bytes 3 files changed, 5161 insertions(+) create mode 100644 lab_4/.gitignore create mode 100644 lab_4/lab4.ipynb create mode 100644 lab_4/requirements.txt diff --git a/lab_4/.gitignore b/lab_4/.gitignore new file mode 100644 index 0000000..6664a32 --- /dev/null +++ b/lab_4/.gitignore @@ -0,0 +1 @@ +/csv/ \ No newline at end of file diff --git a/lab_4/lab4.ipynb b/lab_4/lab4.ipynb new file mode 100644 index 0000000..02d9229 --- /dev/null +++ b/lab_4/lab4.ipynb @@ -0,0 +1,5160 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Лабораторная 4\n", + "Датасет: Набор данных для анализа и прогнозирования сердечного приступа" + ] + }, + { + "cell_type": "code", + "execution_count": 54, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Index(['State', 'Sex', 'GeneralHealth', 'PhysicalHealthDays',\n", + " 'MentalHealthDays', 'LastCheckupTime', 'PhysicalActivities',\n", + " 'SleepHours', 'RemovedTeeth', 'HadHeartAttack', 'HadAngina',\n", + " 'HadStroke', 'HadAsthma', 'HadSkinCancer', 'HadCOPD',\n", + " 'HadDepressiveDisorder', 'HadKidneyDisease', 'HadArthritis',\n", + " 'HadDiabetes', 'DeafOrHardOfHearing', 'BlindOrVisionDifficulty',\n", + " 'DifficultyConcentrating', 'DifficultyWalking',\n", + " 'DifficultyDressingBathing', 'DifficultyErrands', 'SmokerStatus',\n", + " 'ECigaretteUsage', 'ChestScan', 'RaceEthnicityCategory', 'AgeCategory',\n", + " 'HeightInMeters', 'WeightInKilograms', 'BMI', 'AlcoholDrinkers',\n", + " 'HIVTesting', 'FluVaxLast12', 'PneumoVaxEver', 'TetanusLast10Tdap',\n", + " 'HighRiskLastYear', 'CovidPos'],\n", + " dtype='object')\n" + ] + } + ], + "source": [ + "import pandas as pd\n", + "from sklearn import set_config\n", + "\n", + "set_config(transform_output=\"pandas\")\n", + "df = pd.read_csv(\"csv\\\\heart_2022_no_nans.csv\")\n", + "print(df.columns)\n", + "map_heart_disease_to_int = {'No': 0, 'Yes': 1}\n", + "\n", + "TARGET_COLUMN_NAME_CLASSIFICATION = 'HadHeartAttack'\n", + "\n", + "df[TARGET_COLUMN_NAME_CLASSIFICATION] = df[TARGET_COLUMN_NAME_CLASSIFICATION].map(map_heart_disease_to_int).astype('int32')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Классификация" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Бизнес цель 1: \n", + "Предсказание сердечного приступа (HadHeartAttack) на основе других факторов." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Формируем выборки" + ] + }, + { + "cell_type": "code", + "execution_count": 55, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'X_train'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
StateSexGeneralHealthPhysicalHealthDaysMentalHealthDaysLastCheckupTimePhysicalActivitiesSleepHoursRemovedTeethHadHeartAttack...HeightInMetersWeightInKilogramsBMIAlcoholDrinkersHIVTestingFluVaxLast12PneumoVaxEverTetanusLast10TdapHighRiskLastYearCovidPos
6432ArizonaMaleVery good0.05.0Within past 5 years (2 years but less than 5 y...Yes8.0None of them0...1.8877.1121.83YesYesYesNoNo, did not receive any tetanus shot in the pa...NoYes
61767IndianaFemaleVery good0.00.0Within past year (anytime less than 12 months ...Yes6.0None of them0...1.7377.1125.85YesYesNoNoYes, received TdapNoNo
102005MichiganMaleVery good0.00.0Within past year (anytime less than 12 months ...Yes7.0None of them0...1.8583.4624.28YesNoNoNoYes, received TdapNoYes
183791South DakotaFemaleGood10.05.0Within past year (anytime less than 12 months ...Yes7.0None of them0...1.7581.6526.58NoYesNoNoYes, received tetanus shot but not sure what typeNoNo
230656West VirginiaFemaleGood0.00.0Within past year (anytime less than 12 months ...No8.06 or more, but not all0...1.5568.0428.34NoNoNoNoNo, did not receive any tetanus shot in the pa...NoNo
..................................................................
93877MarylandFemaleVery good0.012.0Within past year (anytime less than 12 months ...No6.06 or more, but not all0...1.65113.4041.60NoNoNoNoYes, received TdapNoYes
117856MissouriMaleGood0.00.0Within past year (anytime less than 12 months ...Yes8.01 to 50...1.80117.9336.26NoNoNoNoYes, received tetanus shot but not sure what typeNoYes
41922GeorgiaMaleVery good0.00.0Within past year (anytime less than 12 months ...Yes7.0None of them0...1.78113.4035.87YesNoYesNoYes, received tetanus shot but not sure what typeNoNo
98221MassachusettsFemaleGood5.020.0Within past 2 years (1 year but less than 2 ye...No5.0None of them0...1.7090.7231.32YesYesNoYesYes, received TdapNoNo
151717New YorkMaleVery good2.00.0Within past year (anytime less than 12 months ...Yes7.0None of them0...1.7368.9523.11YesYesYesNoYes, received TdapNoYes
\n", + "

196817 rows × 40 columns

\n", + "
" + ], + "text/plain": [ + " State Sex GeneralHealth PhysicalHealthDays \\\n", + "6432 Arizona Male Very good 0.0 \n", + "61767 Indiana Female Very good 0.0 \n", + "102005 Michigan Male Very good 0.0 \n", + "183791 South Dakota Female Good 10.0 \n", + "230656 West Virginia Female Good 0.0 \n", + "... ... ... ... ... \n", + "93877 Maryland Female Very good 0.0 \n", + "117856 Missouri Male Good 0.0 \n", + "41922 Georgia Male Very good 0.0 \n", + "98221 Massachusetts Female Good 5.0 \n", + "151717 New York Male Very good 2.0 \n", + "\n", + " MentalHealthDays LastCheckupTime \\\n", + "6432 5.0 Within past 5 years (2 years but less than 5 y... \n", + "61767 0.0 Within past year (anytime less than 12 months ... \n", + "102005 0.0 Within past year (anytime less than 12 months ... \n", + "183791 5.0 Within past year (anytime less than 12 months ... \n", + "230656 0.0 Within past year (anytime less than 12 months ... \n", + "... ... ... \n", + "93877 12.0 Within past year (anytime less than 12 months ... \n", + "117856 0.0 Within past year (anytime less than 12 months ... \n", + "41922 0.0 Within past year (anytime less than 12 months ... \n", + "98221 20.0 Within past 2 years (1 year but less than 2 ye... \n", + "151717 0.0 Within past year (anytime less than 12 months ... \n", + "\n", + " PhysicalActivities SleepHours RemovedTeeth HadHeartAttack \\\n", + "6432 Yes 8.0 None of them 0 \n", + "61767 Yes 6.0 None of them 0 \n", + "102005 Yes 7.0 None of them 0 \n", + "183791 Yes 7.0 None of them 0 \n", + "230656 No 8.0 6 or more, but not all 0 \n", + "... ... ... ... ... \n", + "93877 No 6.0 6 or more, but not all 0 \n", + "117856 Yes 8.0 1 to 5 0 \n", + "41922 Yes 7.0 None of them 0 \n", + "98221 No 5.0 None of them 0 \n", + "151717 Yes 7.0 None of them 0 \n", + "\n", + " ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers \\\n", + "6432 ... 1.88 77.11 21.83 Yes \n", + "61767 ... 1.73 77.11 25.85 Yes \n", + "102005 ... 1.85 83.46 24.28 Yes \n", + "183791 ... 1.75 81.65 26.58 No \n", + "230656 ... 1.55 68.04 28.34 No \n", + "... ... ... ... ... ... \n", + "93877 ... 1.65 113.40 41.60 No \n", + "117856 ... 1.80 117.93 36.26 No \n", + "41922 ... 1.78 113.40 35.87 Yes \n", + "98221 ... 1.70 90.72 31.32 Yes \n", + "151717 ... 1.73 68.95 23.11 Yes \n", + "\n", + " HIVTesting FluVaxLast12 PneumoVaxEver \\\n", + "6432 Yes Yes No \n", + "61767 Yes No No \n", + "102005 No No No \n", + "183791 Yes No No \n", + "230656 No No No \n", + "... ... ... ... \n", + "93877 No No No \n", + "117856 No No No \n", + "41922 No Yes No \n", + "98221 Yes No Yes \n", + "151717 Yes Yes No \n", + "\n", + " TetanusLast10Tdap HighRiskLastYear \\\n", + "6432 No, did not receive any tetanus shot in the pa... No \n", + "61767 Yes, received Tdap No \n", + "102005 Yes, received Tdap No \n", + "183791 Yes, received tetanus shot but not sure what type No \n", + "230656 No, did not receive any tetanus shot in the pa... No \n", + "... ... ... \n", + "93877 Yes, received Tdap No \n", + "117856 Yes, received tetanus shot but not sure what type No \n", + "41922 Yes, received tetanus shot but not sure what type No \n", + "98221 Yes, received Tdap No \n", + "151717 Yes, received Tdap No \n", + "\n", + " CovidPos \n", + "6432 Yes \n", + "61767 No \n", + "102005 Yes \n", + "183791 No \n", + "230656 No \n", + "... ... \n", + "93877 Yes \n", + "117856 Yes \n", + "41922 No \n", + "98221 No \n", + "151717 Yes \n", + "\n", + "[196817 rows x 40 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'y_train'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
HadHeartAttack
64320
617670
1020050
1837910
2306560
......
938770
1178560
419220
982210
1517170
\n", + "

196817 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " HadHeartAttack\n", + "6432 0\n", + "61767 0\n", + "102005 0\n", + "183791 0\n", + "230656 0\n", + "... ...\n", + "93877 0\n", + "117856 0\n", + "41922 0\n", + "98221 0\n", + "151717 0\n", + "\n", + "[196817 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'X_test'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
StateSexGeneralHealthPhysicalHealthDaysMentalHealthDaysLastCheckupTimePhysicalActivitiesSleepHoursRemovedTeethHadHeartAttack...HeightInMetersWeightInKilogramsBMIAlcoholDrinkersHIVTestingFluVaxLast12PneumoVaxEverTetanusLast10TdapHighRiskLastYearCovidPos
108080MinnesotaFemaleVery good0.00.0Within past year (anytime less than 12 months ...Yes7.0None of them0...1.6881.6529.05YesNoNoNoYes, received tetanus shot, but not TdapNoYes
109629MinnesotaFemaleVery good1.015.0Within past year (anytime less than 12 months ...Yes6.0None of them0...1.6899.7935.51YesNoYesNoYes, received tetanus shot but not sure what typeNoNo
24640ConnecticutMaleGood15.05.0Within past year (anytime less than 12 months ...Yes7.06 or more, but not all0...1.7072.5725.06YesYesYesYesYes, received tetanus shot but not sure what typeNoNo
12715ArkansasFemaleGood8.030.0Within past year (anytime less than 12 months ...Yes7.01 to 50...1.6386.1832.61YesYesYesNoYes, received TdapNoNo
162549OhioFemaleExcellent0.07.0Within past year (anytime less than 12 months ...Yes4.0None of them0...1.6081.1931.71YesYesYesNoYes, received TdapYesTested positive using home test without a heal...
..................................................................
187130South DakotaMalePoor30.030.0Within past year (anytime less than 12 months ...No4.0None of them0...1.8397.9829.29YesNoNoNoNo, did not receive any tetanus shot in the pa...NoNo
38512FloridaMaleExcellent0.00.0Within past 5 years (2 years but less than 5 y...Yes8.0None of them0...1.83104.3331.19YesNoNoNoYes, received tetanus shot but not sure what typeNoNo
125776NebraskaMaleFair1.02.0Within past year (anytime less than 12 months ...No6.01 to 50...1.7392.9931.17NoYesNoYesYes, received tetanus shot but not sure what typeNoYes
33614FloridaFemaleGood0.00.0Within past year (anytime less than 12 months ...Yes7.0None of them0...1.6065.7725.69YesNoNoYesYes, received TdapNoNo
223067WashingtonMaleExcellent0.02.0Within past 2 years (1 year but less than 2 ye...Yes7.01 to 50...1.7570.0022.86YesYesYesNoYes, received TdapNoNo
\n", + "

49205 rows × 40 columns

\n", + "
" + ], + "text/plain": [ + " State Sex GeneralHealth PhysicalHealthDays \\\n", + "108080 Minnesota Female Very good 0.0 \n", + "109629 Minnesota Female Very good 1.0 \n", + "24640 Connecticut Male Good 15.0 \n", + "12715 Arkansas Female Good 8.0 \n", + "162549 Ohio Female Excellent 0.0 \n", + "... ... ... ... ... \n", + "187130 South Dakota Male Poor 30.0 \n", + "38512 Florida Male Excellent 0.0 \n", + "125776 Nebraska Male Fair 1.0 \n", + "33614 Florida Female Good 0.0 \n", + "223067 Washington Male Excellent 0.0 \n", + "\n", + " MentalHealthDays LastCheckupTime \\\n", + "108080 0.0 Within past year (anytime less than 12 months ... \n", + "109629 15.0 Within past year (anytime less than 12 months ... \n", + "24640 5.0 Within past year (anytime less than 12 months ... \n", + "12715 30.0 Within past year (anytime less than 12 months ... \n", + "162549 7.0 Within past year (anytime less than 12 months ... \n", + "... ... ... \n", + "187130 30.0 Within past year (anytime less than 12 months ... \n", + "38512 0.0 Within past 5 years (2 years but less than 5 y... \n", + "125776 2.0 Within past year (anytime less than 12 months ... \n", + "33614 0.0 Within past year (anytime less than 12 months ... \n", + "223067 2.0 Within past 2 years (1 year but less than 2 ye... \n", + "\n", + " PhysicalActivities SleepHours RemovedTeeth HadHeartAttack \\\n", + "108080 Yes 7.0 None of them 0 \n", + "109629 Yes 6.0 None of them 0 \n", + "24640 Yes 7.0 6 or more, but not all 0 \n", + "12715 Yes 7.0 1 to 5 0 \n", + "162549 Yes 4.0 None of them 0 \n", + "... ... ... ... ... \n", + "187130 No 4.0 None of them 0 \n", + "38512 Yes 8.0 None of them 0 \n", + "125776 No 6.0 1 to 5 0 \n", + "33614 Yes 7.0 None of them 0 \n", + "223067 Yes 7.0 1 to 5 0 \n", + "\n", + " ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers \\\n", + "108080 ... 1.68 81.65 29.05 Yes \n", + "109629 ... 1.68 99.79 35.51 Yes \n", + "24640 ... 1.70 72.57 25.06 Yes \n", + "12715 ... 1.63 86.18 32.61 Yes \n", + "162549 ... 1.60 81.19 31.71 Yes \n", + "... ... ... ... ... ... \n", + "187130 ... 1.83 97.98 29.29 Yes \n", + "38512 ... 1.83 104.33 31.19 Yes \n", + "125776 ... 1.73 92.99 31.17 No \n", + "33614 ... 1.60 65.77 25.69 Yes \n", + "223067 ... 1.75 70.00 22.86 Yes \n", + "\n", + " HIVTesting FluVaxLast12 PneumoVaxEver \\\n", + "108080 No No No \n", + "109629 No Yes No \n", + "24640 Yes Yes Yes \n", + "12715 Yes Yes No \n", + "162549 Yes Yes No \n", + "... ... ... ... \n", + "187130 No No No \n", + "38512 No No No \n", + "125776 Yes No Yes \n", + "33614 No No Yes \n", + "223067 Yes Yes No \n", + "\n", + " TetanusLast10Tdap HighRiskLastYear \\\n", + "108080 Yes, received tetanus shot, but not Tdap No \n", + "109629 Yes, received tetanus shot but not sure what type No \n", + "24640 Yes, received tetanus shot but not sure what type No \n", + "12715 Yes, received Tdap No \n", + "162549 Yes, received Tdap Yes \n", + "... ... ... \n", + "187130 No, did not receive any tetanus shot in the pa... No \n", + "38512 Yes, received tetanus shot but not sure what type No \n", + "125776 Yes, received tetanus shot but not sure what type No \n", + "33614 Yes, received Tdap No \n", + "223067 Yes, received Tdap No \n", + "\n", + " CovidPos \n", + "108080 Yes \n", + "109629 No \n", + "24640 No \n", + "12715 No \n", + "162549 Tested positive using home test without a heal... \n", + "... ... \n", + "187130 No \n", + "38512 No \n", + "125776 Yes \n", + "33614 No \n", + "223067 No \n", + "\n", + "[49205 rows x 40 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'y_test'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
HadHeartAttack
1080800
1096290
246400
127150
1625490
......
1871300
385120
1257760
336140
2230670
\n", + "

49205 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " HadHeartAttack\n", + "108080 0\n", + "109629 0\n", + "24640 0\n", + "12715 0\n", + "162549 0\n", + "... ...\n", + "187130 0\n", + "38512 0\n", + "125776 0\n", + "33614 0\n", + "223067 0\n", + "\n", + "[49205 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from typing import Tuple\n", + "import pandas as pd\n", + "from pandas import DataFrame\n", + "from sklearn.model_selection import train_test_split\n", + "\n", + "def split_stratified_into_train_val_test(\n", + " df_input,\n", + " stratify_colname=\"y\",\n", + " frac_train=0.6,\n", + " frac_val=0.15,\n", + " frac_test=0.25,\n", + " random_state=None,\n", + ") -> Tuple[DataFrame, DataFrame, DataFrame, DataFrame, DataFrame, DataFrame]:\n", + " \n", + " if frac_train + frac_val + frac_test != 1.0:\n", + " raise ValueError(\n", + " \"fractions %f, %f, %f do not add up to 1.0\"\n", + " % (frac_train, frac_val, frac_test)\n", + " )\n", + " if stratify_colname not in df_input.columns:\n", + " raise ValueError(\"%s is not a column in the dataframe\" % (stratify_colname))\n", + " X = df_input # Contains all columns.\n", + " y = df_input[\n", + " [stratify_colname]\n", + " ] # Dataframe of just the column on which to stratify.\n", + " # Split original dataframe into train and temp dataframes.\n", + " df_train, df_temp, y_train, y_temp = train_test_split(\n", + " X, y, stratify=y, test_size=(1.0 - frac_train), random_state=random_state\n", + " )\n", + " if frac_val <= 0:\n", + " assert len(df_input) == len(df_train) + len(df_temp)\n", + " return df_train, pd.DataFrame(), df_temp, y_train, pd.DataFrame(), y_temp\n", + " # Split the temp dataframe into val and test dataframes.\n", + " relative_frac_test = frac_test / (frac_val + frac_test)\n", + " df_val, df_test, y_val, y_test = train_test_split(\n", + " df_temp,\n", + " y_temp,\n", + " stratify=y_temp,\n", + " test_size=relative_frac_test,\n", + " random_state=random_state,\n", + " )\n", + " assert len(df_input) == len(df_train) + len(df_val) + len(df_test)\n", + " return df_train, df_val, df_test, y_train, y_val, y_test\n", + "\n", + "X_train, X_val, X_test, y_train, y_val, y_test = split_stratified_into_train_val_test(\n", + " df, stratify_colname=TARGET_COLUMN_NAME_CLASSIFICATION, frac_train=0.80, frac_val=0, frac_test=0.20, random_state=9\n", + ")\n", + "\n", + "display(\"X_train\", X_train)\n", + "display(\"y_train\", y_train)\n", + "\n", + "display(\"X_test\", X_test)\n", + "display(\"y_test\", y_test)" + ] + }, + { + "cell_type": "code", + "execution_count": 56, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Пропущенные значения по столбцам:\n", + "State 0\n", + "Sex 0\n", + "GeneralHealth 0\n", + "PhysicalHealthDays 0\n", + "MentalHealthDays 0\n", + "LastCheckupTime 0\n", + "PhysicalActivities 0\n", + "SleepHours 0\n", + "RemovedTeeth 0\n", + "HadHeartAttack 0\n", + "HadAngina 0\n", + "HadStroke 0\n", + "HadAsthma 0\n", + "HadSkinCancer 0\n", + "HadCOPD 0\n", + "HadDepressiveDisorder 0\n", + "HadKidneyDisease 0\n", + "HadArthritis 0\n", + "HadDiabetes 0\n", + "DeafOrHardOfHearing 0\n", + "BlindOrVisionDifficulty 0\n", + "DifficultyConcentrating 0\n", + "DifficultyWalking 0\n", + "DifficultyDressingBathing 0\n", + "DifficultyErrands 0\n", + "SmokerStatus 0\n", + "ECigaretteUsage 0\n", + "ChestScan 0\n", + "RaceEthnicityCategory 0\n", + "AgeCategory 0\n", + "HeightInMeters 0\n", + "WeightInKilograms 0\n", + "BMI 0\n", + "AlcoholDrinkers 0\n", + "HIVTesting 0\n", + "FluVaxLast12 0\n", + "PneumoVaxEver 0\n", + "TetanusLast10Tdap 0\n", + "HighRiskLastYear 0\n", + "CovidPos 0\n", + "dtype: int64\n", + "\n", + "Статистический обзор данных:\n", + " PhysicalHealthDays MentalHealthDays SleepHours HadHeartAttack \\\n", + "count 246022.000000 246022.000000 246022.000000 246022.000000 \n", + "mean 4.119026 4.167140 7.021331 0.054609 \n", + "std 8.405844 8.102687 1.440681 0.227216 \n", + "min 0.000000 0.000000 1.000000 0.000000 \n", + "25% 0.000000 0.000000 6.000000 0.000000 \n", + "50% 0.000000 0.000000 7.000000 0.000000 \n", + "75% 3.000000 4.000000 8.000000 0.000000 \n", + "max 30.000000 30.000000 24.000000 1.000000 \n", + "\n", + " HeightInMeters WeightInKilograms BMI \n", + "count 246022.000000 246022.000000 246022.000000 \n", + "mean 1.705150 83.615179 28.668136 \n", + "std 0.106654 21.323156 6.513973 \n", + "min 0.910000 28.120000 12.020000 \n", + "25% 1.630000 68.040000 24.270000 \n", + "50% 1.700000 81.650000 27.460000 \n", + "75% 1.780000 95.250000 31.890000 \n", + "max 2.410000 292.570000 97.650000 \n" + ] + } + ], + "source": [ + "null_values = df.isnull().sum()\n", + "print(\"Пропущенные значения по столбцам:\")\n", + "print(null_values)\n", + "\n", + "stat_summary = df.describe()\n", + "print(\"\\nСтатистический обзор данных:\")\n", + "print(stat_summary)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Формируем конвеер для классификации данных и проверка конвеера" + ] + }, + { + "cell_type": "code", + "execution_count": 57, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
PhysicalHealthDaysMentalHealthDaysSleepHoursHadHeartAttackHeightInMetersWeightInKilogramsBMIState_AlaskaState_ArizonaState_Arkansas...AlcoholDrinkers_YesHIVTesting_YesFluVaxLast12_YesPneumoVaxEver_YesTetanusLast10Tdap_Yes, received TdapTetanusLast10Tdap_Yes, received tetanus shot but not sure what typeTetanusLast10Tdap_Yes, received tetanus shot, but not TdapHighRiskLastYear_YesCovidPos_Tested positive using home test without a health professionalCovidPos_Yes
6432-0.4901790.1031240.677965-0.240341.639362-0.304540-1.0513140.01.00.0...1.01.01.00.00.00.00.00.00.01.0
61767-0.490179-0.513985-0.708460-0.240340.233664-0.304540-0.4329660.00.00.0...1.01.00.00.01.00.00.00.00.00.0
102005-0.490179-0.513985-0.015247-0.240341.358222-0.006656-0.6744600.00.00.0...1.00.00.00.01.00.00.00.00.01.0
1837910.6990480.103124-0.015247-0.240340.421091-0.091564-0.3206780.00.00.0...0.01.00.00.00.01.00.00.00.00.0
230656-0.490179-0.5139850.677965-0.24034-1.453173-0.730021-0.0499590.00.00.0...0.00.00.00.00.00.00.00.00.00.0
..................................................................
93877-0.4901790.967076-0.708460-0.24034-0.5160411.3978561.9896660.00.00.0...0.00.00.00.01.00.00.00.00.01.0
117856-0.490179-0.5139850.677965-0.240340.8896561.6103621.1682790.00.00.0...0.00.00.00.00.01.00.00.00.01.0
41922-0.490179-0.513985-0.015247-0.240340.7022301.3978561.1082900.00.00.0...1.00.01.00.00.01.00.00.00.00.0
982210.1044351.954450-1.401672-0.24034-0.0474750.3339170.4084180.00.00.0...1.01.00.01.01.00.00.00.00.00.0
151717-0.252334-0.513985-0.015247-0.240340.233664-0.687332-0.8544270.00.00.0...1.01.01.00.01.00.00.00.00.01.0
\n", + "

196817 rows × 109 columns

\n", + "
" + ], + "text/plain": [ + " PhysicalHealthDays MentalHealthDays SleepHours HadHeartAttack \\\n", + "6432 -0.490179 0.103124 0.677965 -0.24034 \n", + "61767 -0.490179 -0.513985 -0.708460 -0.24034 \n", + "102005 -0.490179 -0.513985 -0.015247 -0.24034 \n", + "183791 0.699048 0.103124 -0.015247 -0.24034 \n", + "230656 -0.490179 -0.513985 0.677965 -0.24034 \n", + "... ... ... ... ... \n", + "93877 -0.490179 0.967076 -0.708460 -0.24034 \n", + "117856 -0.490179 -0.513985 0.677965 -0.24034 \n", + "41922 -0.490179 -0.513985 -0.015247 -0.24034 \n", + "98221 0.104435 1.954450 -1.401672 -0.24034 \n", + "151717 -0.252334 -0.513985 -0.015247 -0.24034 \n", + "\n", + " HeightInMeters WeightInKilograms BMI State_Alaska \\\n", + "6432 1.639362 -0.304540 -1.051314 0.0 \n", + "61767 0.233664 -0.304540 -0.432966 0.0 \n", + "102005 1.358222 -0.006656 -0.674460 0.0 \n", + "183791 0.421091 -0.091564 -0.320678 0.0 \n", + "230656 -1.453173 -0.730021 -0.049959 0.0 \n", + "... ... ... ... ... \n", + "93877 -0.516041 1.397856 1.989666 0.0 \n", + "117856 0.889656 1.610362 1.168279 0.0 \n", + "41922 0.702230 1.397856 1.108290 0.0 \n", + "98221 -0.047475 0.333917 0.408418 0.0 \n", + "151717 0.233664 -0.687332 -0.854427 0.0 \n", + "\n", + " State_Arizona State_Arkansas ... AlcoholDrinkers_Yes \\\n", + "6432 1.0 0.0 ... 1.0 \n", + "61767 0.0 0.0 ... 1.0 \n", + "102005 0.0 0.0 ... 1.0 \n", + "183791 0.0 0.0 ... 0.0 \n", + "230656 0.0 0.0 ... 0.0 \n", + "... ... ... ... ... \n", + "93877 0.0 0.0 ... 0.0 \n", + "117856 0.0 0.0 ... 0.0 \n", + "41922 0.0 0.0 ... 1.0 \n", + "98221 0.0 0.0 ... 1.0 \n", + "151717 0.0 0.0 ... 1.0 \n", + "\n", + " HIVTesting_Yes FluVaxLast12_Yes PneumoVaxEver_Yes \\\n", + "6432 1.0 1.0 0.0 \n", + "61767 1.0 0.0 0.0 \n", + "102005 0.0 0.0 0.0 \n", + "183791 1.0 0.0 0.0 \n", + "230656 0.0 0.0 0.0 \n", + "... ... ... ... \n", + "93877 0.0 0.0 0.0 \n", + "117856 0.0 0.0 0.0 \n", + "41922 0.0 1.0 0.0 \n", + "98221 1.0 0.0 1.0 \n", + "151717 1.0 1.0 0.0 \n", + "\n", + " TetanusLast10Tdap_Yes, received Tdap \\\n", + "6432 0.0 \n", + "61767 1.0 \n", + "102005 1.0 \n", + "183791 0.0 \n", + "230656 0.0 \n", + "... ... \n", + "93877 1.0 \n", + "117856 0.0 \n", + "41922 0.0 \n", + "98221 1.0 \n", + "151717 1.0 \n", + "\n", + " TetanusLast10Tdap_Yes, received tetanus shot but not sure what type \\\n", + "6432 0.0 \n", + "61767 0.0 \n", + "102005 0.0 \n", + "183791 1.0 \n", + "230656 0.0 \n", + "... ... \n", + "93877 0.0 \n", + "117856 1.0 \n", + "41922 1.0 \n", + "98221 0.0 \n", + "151717 0.0 \n", + "\n", + " TetanusLast10Tdap_Yes, received tetanus shot, but not Tdap \\\n", + "6432 0.0 \n", + "61767 0.0 \n", + "102005 0.0 \n", + "183791 0.0 \n", + "230656 0.0 \n", + "... ... \n", + "93877 0.0 \n", + "117856 0.0 \n", + "41922 0.0 \n", + "98221 0.0 \n", + "151717 0.0 \n", + "\n", + " HighRiskLastYear_Yes \\\n", + "6432 0.0 \n", + "61767 0.0 \n", + "102005 0.0 \n", + "183791 0.0 \n", + "230656 0.0 \n", + "... ... \n", + "93877 0.0 \n", + "117856 0.0 \n", + "41922 0.0 \n", + "98221 0.0 \n", + "151717 0.0 \n", + "\n", + " CovidPos_Tested positive using home test without a health professional \\\n", + "6432 0.0 \n", + "61767 0.0 \n", + "102005 0.0 \n", + "183791 0.0 \n", + "230656 0.0 \n", + "... ... \n", + "93877 0.0 \n", + "117856 0.0 \n", + "41922 0.0 \n", + "98221 0.0 \n", + "151717 0.0 \n", + "\n", + " CovidPos_Yes \n", + "6432 1.0 \n", + "61767 0.0 \n", + "102005 1.0 \n", + "183791 0.0 \n", + "230656 0.0 \n", + "... ... \n", + "93877 1.0 \n", + "117856 1.0 \n", + "41922 0.0 \n", + "98221 0.0 \n", + "151717 1.0 \n", + "\n", + "[196817 rows x 109 columns]" + ] + }, + "execution_count": 57, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from sklearn.compose import ColumnTransformer\n", + "from sklearn.discriminant_analysis import StandardScaler\n", + "from sklearn.impute import SimpleImputer\n", + "from sklearn.pipeline import Pipeline\n", + "from sklearn.preprocessing import OneHotEncoder\n", + "\n", + "columns_to_drop = ['AgeCategory', 'Sex']\n", + "num_columns = [\n", + " column\n", + " for column in df.columns\n", + " if column not in columns_to_drop and df[column].dtype != \"object\"\n", + "]\n", + "cat_columns = [\n", + " column\n", + " for column in df.columns\n", + " if column not in columns_to_drop and df[column].dtype == \"object\"\n", + "]\n", + "\n", + "num_imputer = SimpleImputer(strategy=\"median\")\n", + "num_scaler = StandardScaler()\n", + "preprocessing_num = Pipeline(\n", + " [\n", + " (\"imputer\", num_imputer),\n", + " (\"scaler\", num_scaler),\n", + " ]\n", + ")\n", + "\n", + "cat_imputer = SimpleImputer(strategy=\"constant\", fill_value=\"unknown\")\n", + "cat_encoder = OneHotEncoder(handle_unknown=\"ignore\", sparse_output=False, drop=\"first\")\n", + "preprocessing_cat = Pipeline(\n", + " [\n", + " (\"imputer\", cat_imputer),\n", + " (\"encoder\", cat_encoder),\n", + " ]\n", + ")\n", + "\n", + "features_preprocessing = ColumnTransformer(\n", + " verbose_feature_names_out=False,\n", + " transformers=[\n", + " (\"prepocessing_num\", preprocessing_num, num_columns),\n", + " (\"prepocessing_cat\", preprocessing_cat, cat_columns),\n", + " ],\n", + " remainder=\"passthrough\"\n", + ")\n", + "\n", + "drop_columns = ColumnTransformer(\n", + " verbose_feature_names_out=False,\n", + " transformers=[\n", + " (\"drop_columns\", \"drop\", columns_to_drop),\n", + " ],\n", + " remainder=\"passthrough\",\n", + ")\n", + "\n", + "\n", + "pipeline_end = Pipeline(\n", + " [\n", + " (\"features_preprocessing\", features_preprocessing),\n", + " (\"drop_columns\", drop_columns),\n", + " ]\n", + ")\n", + "\n", + "preprocessing_result = pipeline_end.fit_transform(X_train)\n", + "preprocessed_df = pd.DataFrame(\n", + " preprocessing_result,\n", + " columns=pipeline_end.get_feature_names_out(),\n", + ")\n", + "\n", + "preprocessed_df" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Формируем набор моделей" + ] + }, + { + "cell_type": "code", + "execution_count": 58, + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn import ensemble, linear_model, naive_bayes, neighbors, neural_network, tree\n", + "\n", + "\n", + "class_models = {\n", + " \"logistic\": {\"model\": linear_model.LogisticRegression()},\n", + " \"ridge\": {\"model\": linear_model.LogisticRegression(penalty=\"l2\", class_weight=\"balanced\")},\n", + " \"decision_tree\": {\n", + " \"model\": tree.DecisionTreeClassifier(max_depth=7, random_state=9)\n", + " },\n", + " \"knn\": {\"model\": neighbors.KNeighborsClassifier(n_neighbors=7)},\n", + " \"naive_bayes\": {\"model\": naive_bayes.GaussianNB()},\n", + " \"gradient_boosting\": {\n", + " \"model\": ensemble.GradientBoostingClassifier(n_estimators=210)\n", + " },\n", + " \"random_forest\": {\n", + " \"model\": ensemble.RandomForestClassifier(\n", + " max_depth=11, class_weight=\"balanced\", random_state=9\n", + " )\n", + " },\n", + " \"mlp\": {\n", + " \"model\": neural_network.MLPClassifier(\n", + " hidden_layer_sizes=(7,),\n", + " max_iter=500,\n", + " early_stopping=True,\n", + " random_state=9,\n", + " )\n", + " },\n", + "}" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Обучаем модели и тестируем их" + ] + }, + { + "cell_type": "code", + "execution_count": 59, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Model: logistic\n", + "Model: ridge\n", + "Model: decision_tree\n", + "Model: knn\n", + "Model: naive_bayes\n", + "Model: gradient_boosting\n", + "Model: random_forest\n", + "Model: mlp\n" + ] + } + ], + "source": [ + "import numpy as np\n", + "from sklearn import metrics\n", + "\n", + "for model_name in class_models.keys():\n", + " print(f\"Model: {model_name}\")\n", + " model = class_models[model_name][\"model\"]\n", + "\n", + " model_pipeline = Pipeline([(\"pipeline\", pipeline_end), (\"model\", model)])\n", + " model_pipeline = model_pipeline.fit(X_train, y_train.values.ravel())\n", + "\n", + " y_train_predict = model_pipeline.predict(X_train)\n", + " y_test_probs = model_pipeline.predict_proba(X_test)[:, 1]\n", + " y_test_predict = np.where(y_test_probs > 0.5, 1, 0)\n", + "\n", + " class_models[model_name][\"pipeline\"] = model_pipeline\n", + " class_models[model_name][\"probs\"] = y_test_probs\n", + " class_models[model_name][\"preds\"] = y_test_predict\n", + "\n", + " class_models[model_name][\"Precision_train\"] = metrics.precision_score(\n", + " y_train, y_train_predict\n", + " )\n", + " class_models[model_name][\"Precision_test\"] = metrics.precision_score(\n", + " y_test, y_test_predict\n", + " )\n", + " class_models[model_name][\"Recall_train\"] = metrics.recall_score(\n", + " y_train, y_train_predict\n", + " )\n", + " class_models[model_name][\"Recall_test\"] = metrics.recall_score(\n", + " y_test, y_test_predict\n", + " )\n", + " class_models[model_name][\"Accuracy_train\"] = metrics.accuracy_score(\n", + " y_train, y_train_predict\n", + " )\n", + " class_models[model_name][\"Accuracy_test\"] = metrics.accuracy_score(\n", + " y_test, y_test_predict\n", + " )\n", + " class_models[model_name][\"ROC_AUC_test\"] = metrics.roc_auc_score(\n", + " y_test, y_test_probs\n", + " )\n", + " class_models[model_name][\"F1_train\"] = metrics.f1_score(y_train, y_train_predict, average=None)\n", + " class_models[model_name][\"F1_test\"] = metrics.f1_score(y_test, y_test_predict, average=None)\n", + " class_models[model_name][\"MCC_test\"] = metrics.matthews_corrcoef(\n", + " y_test, y_test_predict\n", + " )\n", + " class_models[model_name][\"Cohen_kappa_test\"] = metrics.cohen_kappa_score(\n", + " y_test, y_test_predict\n", + " )\n", + " class_models[model_name][\"Confusion_matrix\"] = metrics.confusion_matrix(\n", + " y_test, y_test_predict\n", + " )" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Матрица неточностей" + ] + }, + { + "cell_type": "code", + "execution_count": 60, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from matplotlib import pyplot as plt\n", + "from sklearn.metrics import ConfusionMatrixDisplay\n", + "\n", + "_, ax = plt.subplots(int(len(class_models) / 2), 2, figsize=(12, 10), sharex=False, sharey=False)\n", + "for index, key in enumerate(class_models.keys()):\n", + " c_matrix = class_models[key][\"Confusion_matrix\"]\n", + " disp = ConfusionMatrixDisplay(\n", + " confusion_matrix=c_matrix, display_labels=[f\"No {TARGET_COLUMN_NAME_CLASSIFICATION}\", TARGET_COLUMN_NAME_CLASSIFICATION]\n", + " ).plot(ax=ax.flat[index])\n", + " disp.ax_.set_title(key)\n", + "\n", + "plt.subplots_adjust(top=1, bottom=0, hspace=0.4, wspace=0.1)\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Точность, полнота, верность (аккуратность), F-мера" + ] + }, + { + "cell_type": "code", + "execution_count": 61, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Precision_trainPrecision_testRecall_trainRecall_testAccuracy_trainAccuracy_testF1_trainF1_test
logistic1.01.01.0000001.01.0000001.0[1.0, 1.0][1.0, 1.0]
ridge1.01.01.0000001.01.0000001.0[1.0, 1.0][1.0, 1.0]
decision_tree1.01.01.0000001.01.0000001.0[1.0, 1.0][1.0, 1.0]
knn1.01.00.9999071.00.9999951.0[0.9999973128320332, 0.9999534775529193][1.0, 1.0]
naive_bayes1.01.01.0000001.01.0000001.0[1.0, 1.0][1.0, 1.0]
gradient_boosting1.01.01.0000001.01.0000001.0[1.0, 1.0][1.0, 1.0]
random_forest1.01.01.0000001.01.0000001.0[1.0, 1.0][1.0, 1.0]
mlp1.01.01.0000001.01.0000001.0[1.0, 1.0][1.0, 1.0]
\n", + "
" + ], + "text/plain": [ + " Precision_train Precision_test Recall_train Recall_test \\\n", + "logistic 1.0 1.0 1.000000 1.0 \n", + "ridge 1.0 1.0 1.000000 1.0 \n", + "decision_tree 1.0 1.0 1.000000 1.0 \n", + "knn 1.0 1.0 0.999907 1.0 \n", + "naive_bayes 1.0 1.0 1.000000 1.0 \n", + "gradient_boosting 1.0 1.0 1.000000 1.0 \n", + "random_forest 1.0 1.0 1.000000 1.0 \n", + "mlp 1.0 1.0 1.000000 1.0 \n", + "\n", + " Accuracy_train Accuracy_test \\\n", + "logistic 1.000000 1.0 \n", + "ridge 1.000000 1.0 \n", + "decision_tree 1.000000 1.0 \n", + "knn 0.999995 1.0 \n", + "naive_bayes 1.000000 1.0 \n", + "gradient_boosting 1.000000 1.0 \n", + "random_forest 1.000000 1.0 \n", + "mlp 1.000000 1.0 \n", + "\n", + " F1_train F1_test \n", + "logistic [1.0, 1.0] [1.0, 1.0] \n", + "ridge [1.0, 1.0] [1.0, 1.0] \n", + "decision_tree [1.0, 1.0] [1.0, 1.0] \n", + "knn [0.9999973128320332, 0.9999534775529193] [1.0, 1.0] \n", + "naive_bayes [1.0, 1.0] [1.0, 1.0] \n", + "gradient_boosting [1.0, 1.0] [1.0, 1.0] \n", + "random_forest [1.0, 1.0] [1.0, 1.0] \n", + "mlp [1.0, 1.0] [1.0, 1.0] " + ] + }, + "execution_count": 61, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "class_metrics = pd.DataFrame.from_dict(class_models, \"index\")[\n", + " [\n", + " \"Precision_train\",\n", + " \"Precision_test\",\n", + " \"Recall_train\",\n", + " \"Recall_test\",\n", + " \"Accuracy_train\",\n", + " \"Accuracy_test\",\n", + " \"F1_train\",\n", + " \"F1_test\",\n", + " ]\n", + "]\n", + "class_metrics.sort_values(\n", + " by=\"Accuracy_test\", ascending=False\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "ROC-кривая, каппа Коэна, коэффициент корреляции Мэтьюса" + ] + }, + { + "cell_type": "code", + "execution_count": 62, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Accuracy_testF1_testROC_AUC_testCohen_kappa_testMCC_test
logistic1.0[1.0, 1.0]1.01.01.0
ridge1.0[1.0, 1.0]1.01.01.0
decision_tree1.0[1.0, 1.0]1.01.01.0
knn1.0[1.0, 1.0]1.01.01.0
naive_bayes1.0[1.0, 1.0]1.01.01.0
gradient_boosting1.0[1.0, 1.0]1.01.01.0
random_forest1.0[1.0, 1.0]1.01.01.0
mlp1.0[1.0, 1.0]1.01.01.0
\n", + "
" + ], + "text/plain": [ + " Accuracy_test F1_test ROC_AUC_test Cohen_kappa_test \\\n", + "logistic 1.0 [1.0, 1.0] 1.0 1.0 \n", + "ridge 1.0 [1.0, 1.0] 1.0 1.0 \n", + "decision_tree 1.0 [1.0, 1.0] 1.0 1.0 \n", + "knn 1.0 [1.0, 1.0] 1.0 1.0 \n", + "naive_bayes 1.0 [1.0, 1.0] 1.0 1.0 \n", + "gradient_boosting 1.0 [1.0, 1.0] 1.0 1.0 \n", + "random_forest 1.0 [1.0, 1.0] 1.0 1.0 \n", + "mlp 1.0 [1.0, 1.0] 1.0 1.0 \n", + "\n", + " MCC_test \n", + "logistic 1.0 \n", + "ridge 1.0 \n", + "decision_tree 1.0 \n", + "knn 1.0 \n", + "naive_bayes 1.0 \n", + "gradient_boosting 1.0 \n", + "random_forest 1.0 \n", + "mlp 1.0 " + ] + }, + "execution_count": 62, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "class_metrics = pd.DataFrame.from_dict(class_models, \"index\")[\n", + " [\n", + " \"Accuracy_test\",\n", + " \"F1_test\",\n", + " \"ROC_AUC_test\",\n", + " \"Cohen_kappa_test\",\n", + " \"MCC_test\",\n", + " ]\n", + "]\n", + "class_metrics.sort_values(by=\"ROC_AUC_test\", ascending=False)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Лучшая модель" + ] + }, + { + "cell_type": "code", + "execution_count": 63, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'logistic'" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "best_model = str(class_metrics.sort_values(by=\"MCC_test\", ascending=False).iloc[0].name)\n", + "\n", + "display(best_model)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Находим ошибки" + ] + }, + { + "cell_type": "code", + "execution_count": 64, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'Error items count: 0'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
StatePredictedSexGeneralHealthPhysicalHealthDaysMentalHealthDaysLastCheckupTimePhysicalActivitiesSleepHoursRemovedTeeth...HeightInMetersWeightInKilogramsBMIAlcoholDrinkersHIVTestingFluVaxLast12PneumoVaxEverTetanusLast10TdapHighRiskLastYearCovidPos
\n", + "

0 rows × 41 columns

\n", + "
" + ], + "text/plain": [ + "Empty DataFrame\n", + "Columns: [State, Predicted, Sex, GeneralHealth, PhysicalHealthDays, MentalHealthDays, LastCheckupTime, PhysicalActivities, SleepHours, RemovedTeeth, HadHeartAttack, HadAngina, HadStroke, HadAsthma, HadSkinCancer, HadCOPD, HadDepressiveDisorder, HadKidneyDisease, HadArthritis, HadDiabetes, DeafOrHardOfHearing, BlindOrVisionDifficulty, DifficultyConcentrating, DifficultyWalking, DifficultyDressingBathing, DifficultyErrands, SmokerStatus, ECigaretteUsage, ChestScan, RaceEthnicityCategory, AgeCategory, HeightInMeters, WeightInKilograms, BMI, AlcoholDrinkers, HIVTesting, FluVaxLast12, PneumoVaxEver, TetanusLast10Tdap, HighRiskLastYear, CovidPos]\n", + "Index: []\n", + "\n", + "[0 rows x 41 columns]" + ] + }, + "execution_count": 64, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "preprocessing_result = pipeline_end.transform(X_test)\n", + "preprocessed_df = pd.DataFrame(\n", + " preprocessing_result,\n", + " columns=pipeline_end.get_feature_names_out(),\n", + ")\n", + "\n", + "y_new_pred = class_models[best_model][\"preds\"]\n", + "\n", + "error_index = y_test[y_test[TARGET_COLUMN_NAME_CLASSIFICATION] != y_new_pred].index.tolist()\n", + "display(f\"Error items count: {len(error_index)}\")\n", + "\n", + "error_predicted = pd.Series(y_new_pred, index=y_test.index).loc[error_index]\n", + "error_df = X_test.loc[error_index].copy()\n", + "error_df.insert(loc=1, column=\"Predicted\", value=error_predicted)\n", + "error_df.sort_index()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Пример использования модели (конвейера) для предсказания" + ] + }, + { + "cell_type": "code", + "execution_count": 65, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
StateSexGeneralHealthPhysicalHealthDaysMentalHealthDaysLastCheckupTimePhysicalActivitiesSleepHoursRemovedTeethHadHeartAttack...HeightInMetersWeightInKilogramsBMIAlcoholDrinkersHIVTestingFluVaxLast12PneumoVaxEverTetanusLast10TdapHighRiskLastYearCovidPos
187130South DakotaMalePoor30.030.0Within past year (anytime less than 12 months ...No4.0None of them0...1.8397.9829.29YesNoNoNoNo, did not receive any tetanus shot in the pa...NoNo
\n", + "

1 rows × 40 columns

\n", + "
" + ], + "text/plain": [ + " State Sex GeneralHealth PhysicalHealthDays MentalHealthDays \\\n", + "187130 South Dakota Male Poor 30.0 30.0 \n", + "\n", + " LastCheckupTime PhysicalActivities \\\n", + "187130 Within past year (anytime less than 12 months ... No \n", + "\n", + " SleepHours RemovedTeeth HadHeartAttack ... HeightInMeters \\\n", + "187130 4.0 None of them 0 ... 1.83 \n", + "\n", + " WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 \\\n", + "187130 97.98 29.29 Yes No No \n", + "\n", + " PneumoVaxEver TetanusLast10Tdap \\\n", + "187130 No No, did not receive any tetanus shot in the pa... \n", + "\n", + " HighRiskLastYear CovidPos \n", + "187130 No No \n", + "\n", + "[1 rows x 40 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
PhysicalHealthDaysMentalHealthDaysSleepHoursHadHeartAttackHeightInMetersWeightInKilogramsBMIState_AlaskaState_ArizonaState_Arkansas...AlcoholDrinkers_YesHIVTesting_YesFluVaxLast12_YesPneumoVaxEver_YesTetanusLast10Tdap_Yes, received TdapTetanusLast10Tdap_Yes, received tetanus shot but not sure what typeTetanusLast10Tdap_Yes, received tetanus shot, but not TdapHighRiskLastYear_YesCovidPos_Tested positive using home test without a health professionalCovidPos_Yes
1871303.0775033.188668-2.094884-0.240341.1707960.674490.0961680.00.00.0...1.00.00.00.00.00.00.00.00.00.0
\n", + "

1 rows × 109 columns

\n", + "
" + ], + "text/plain": [ + " PhysicalHealthDays MentalHealthDays SleepHours HadHeartAttack \\\n", + "187130 3.077503 3.188668 -2.094884 -0.24034 \n", + "\n", + " HeightInMeters WeightInKilograms BMI State_Alaska \\\n", + "187130 1.170796 0.67449 0.096168 0.0 \n", + "\n", + " State_Arizona State_Arkansas ... AlcoholDrinkers_Yes \\\n", + "187130 0.0 0.0 ... 1.0 \n", + "\n", + " HIVTesting_Yes FluVaxLast12_Yes PneumoVaxEver_Yes \\\n", + "187130 0.0 0.0 0.0 \n", + "\n", + " TetanusLast10Tdap_Yes, received Tdap \\\n", + "187130 0.0 \n", + "\n", + " TetanusLast10Tdap_Yes, received tetanus shot but not sure what type \\\n", + "187130 0.0 \n", + "\n", + " TetanusLast10Tdap_Yes, received tetanus shot, but not Tdap \\\n", + "187130 0.0 \n", + "\n", + " HighRiskLastYear_Yes \\\n", + "187130 0.0 \n", + "\n", + " CovidPos_Tested positive using home test without a health professional \\\n", + "187130 0.0 \n", + "\n", + " CovidPos_Yes \n", + "187130 0.0 \n", + "\n", + "[1 rows x 109 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'predicted: 0 (proba: [9.99540301e-01 4.59698535e-04])'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'real: 0'" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "model = class_models[best_model][\"pipeline\"]\n", + "\n", + "\n", + "example_id = 187130\n", + "test = pd.DataFrame(X_test.loc[example_id, :]).T\n", + "test_preprocessed = pd.DataFrame(preprocessed_df.loc[example_id, :]).T\n", + "display(test)\n", + "display(test_preprocessed)\n", + "result_proba = model.predict_proba(test)[0]\n", + "result = model.predict(test)[0]\n", + "real = int(y_test.loc[example_id].values[0])\n", + "display(f\"predicted: {result} (proba: {result_proba})\")\n", + "display(f\"real: {real}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Создаем гиперпараметры методом поиска по сетке" + ] + }, + { + "cell_type": "code", + "execution_count": 66, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "d:\\code\\AIM-PIbd-31-Potapov-N-S\\lab_4\\.venv\\Lib\\site-packages\\numpy\\ma\\core.py:2881: RuntimeWarning: invalid value encountered in cast\n", + " _data = np.array(data, dtype=dtype, copy=copy,\n" + ] + }, + { + "data": { + "text/plain": [ + "{'model__criterion': 'gini',\n", + " 'model__max_depth': 10,\n", + " 'model__max_features': 'sqrt',\n", + " 'model__n_estimators': 100}" + ] + }, + "execution_count": 66, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from sklearn.model_selection import GridSearchCV\n", + "\n", + "\n", + "optimized_model_type = 'random_forest'\n", + "random_state = 9\n", + "\n", + "random_forest_model = class_models[optimized_model_type][\"pipeline\"]\n", + "\n", + "param_grid = {\n", + " \"model__n_estimators\": [10, 50, 100],\n", + " \"model__max_features\": [\"sqrt\", \"log2\"],\n", + " \"model__max_depth\": [5, 7, 10],\n", + " \"model__criterion\": [\"gini\", \"entropy\"],\n", + "}\n", + "\n", + "gs_optomizer = GridSearchCV(\n", + " estimator=random_forest_model, param_grid=param_grid, n_jobs=-1\n", + ")\n", + "gs_optomizer.fit(X_train, y_train.values.ravel())\n", + "gs_optomizer.best_params_\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Обучение модели с новыми гиперпараметрами" + ] + }, + { + "cell_type": "code", + "execution_count": 67, + "metadata": {}, + "outputs": [], + "source": [ + "optimized_model = ensemble.RandomForestClassifier(\n", + " random_state=42,\n", + " criterion=\"gini\",\n", + " max_depth=5,\n", + " max_features=\"sqrt\",\n", + " n_estimators=50,\n", + ")\n", + "\n", + "result = {}\n", + "\n", + "result[\"pipeline\"] = Pipeline([(\"pipeline\", pipeline_end), (\"model\", optimized_model)]).fit(X_train, y_train.values.ravel())\n", + "result[\"train_preds\"] = result[\"pipeline\"].predict(X_train)\n", + "result[\"probs\"] = result[\"pipeline\"].predict_proba(X_test)[:, 1]\n", + "result[\"preds\"] = np.where(result[\"probs\"] > 0.5, 1, 0)\n", + "\n", + "result[\"Precision_train\"] = metrics.precision_score(y_train, result[\"train_preds\"])\n", + "result[\"Precision_test\"] = metrics.precision_score(y_test, result[\"preds\"])\n", + "result[\"Recall_train\"] = metrics.recall_score(y_train, result[\"train_preds\"])\n", + "result[\"Recall_test\"] = metrics.recall_score(y_test, result[\"preds\"])\n", + "result[\"Accuracy_train\"] = metrics.accuracy_score(y_train, result[\"train_preds\"])\n", + "result[\"Accuracy_test\"] = metrics.accuracy_score(y_test, result[\"preds\"])\n", + "result[\"ROC_AUC_test\"] = metrics.roc_auc_score(y_test, result[\"probs\"])\n", + "result[\"F1_train\"] = metrics.f1_score(y_train, result[\"train_preds\"])\n", + "result[\"F1_test\"] = metrics.f1_score(y_test, result[\"preds\"])\n", + "result[\"MCC_test\"] = metrics.matthews_corrcoef(y_test, result[\"preds\"])\n", + "result[\"Cohen_kappa_test\"] = metrics.cohen_kappa_score(y_test, result[\"preds\"])\n", + "result[\"Confusion_matrix\"] = metrics.confusion_matrix(y_test, result[\"preds\"])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Формирование данных для оценки старой и новой версии модели и сама оценка данных" + ] + }, + { + "cell_type": "code", + "execution_count": 68, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Precision_trainPrecision_testRecall_trainRecall_testAccuracy_trainAccuracy_testF1_trainF1_test
Name
Old1.01.01.01.01.01.0[1.0, 1.0][1.0, 1.0]
New1.01.00.3049870.2988460.9620460.9617110.4674180.460172
\n", + "
" + ], + "text/plain": [ + " Precision_train Precision_test Recall_train Recall_test Accuracy_train \\\n", + "Name \n", + "Old 1.0 1.0 1.0 1.0 1.0 \n", + "New 1.0 1.0 0.304987 0.298846 0.962046 \n", + "\n", + " Accuracy_test F1_train F1_test \n", + "Name \n", + "Old 1.0 [1.0, 1.0] [1.0, 1.0] \n", + "New 0.961711 0.467418 0.460172 " + ] + }, + "execution_count": 68, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "optimized_metrics = pd.DataFrame(columns=list(result.keys()))\n", + "optimized_metrics.loc[len(optimized_metrics)] = pd.Series(\n", + " data=class_models[optimized_model_type]\n", + ")\n", + "optimized_metrics.loc[len(optimized_metrics)] = pd.Series(\n", + " data=result\n", + ")\n", + "optimized_metrics.insert(loc=0, column=\"Name\", value=[\"Old\", \"New\"])\n", + "optimized_metrics = optimized_metrics.set_index(\"Name\")\n", + "\n", + "optimized_metrics[\n", + " [\n", + " \"Precision_train\",\n", + " \"Precision_test\",\n", + " \"Recall_train\",\n", + " \"Recall_test\",\n", + " \"Accuracy_train\",\n", + " \"Accuracy_test\",\n", + " \"F1_train\",\n", + " \"F1_test\",\n", + " ]\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": 69, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Accuracy_testF1_testROC_AUC_testCohen_kappa_testMCC_test
Name
Old1.0[1.0, 1.0]1.01.01.0
New0.9617110.4601720.9999940.4462570.535924
\n", + "
" + ], + "text/plain": [ + " Accuracy_test F1_test ROC_AUC_test Cohen_kappa_test MCC_test\n", + "Name \n", + "Old 1.0 [1.0, 1.0] 1.0 1.0 1.0\n", + "New 0.961711 0.460172 0.999994 0.446257 0.535924" + ] + }, + "execution_count": 69, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "optimized_metrics[\n", + " [\n", + " \"Accuracy_test\",\n", + " \"F1_test\",\n", + " \"ROC_AUC_test\",\n", + " \"Cohen_kappa_test\",\n", + " \"MCC_test\",\n", + " ]\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": 70, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "_, ax = plt.subplots(1, 2, figsize=(10, 4), sharex=False, sharey=False\n", + ")\n", + "\n", + "for index in range(0, len(optimized_metrics)):\n", + " c_matrix = optimized_metrics.iloc[index][\"Confusion_matrix\"]\n", + " disp = ConfusionMatrixDisplay(\n", + " confusion_matrix=c_matrix, display_labels=[f\"No {TARGET_COLUMN_NAME_CLASSIFICATION}\", TARGET_COLUMN_NAME_CLASSIFICATION]\n", + " ).plot(ax=ax.flat[index])\n", + "\n", + "plt.subplots_adjust(top=1, bottom=0, hspace=0.4, wspace=0.3)\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Модель хорошо классифицировала объекты, которые относятся к \"No HadHeartAttack\" и \"HadHeartAttack\"" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Регрессия" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Бизнес цель 2: \n", + "Предсказание среднего количества часов сна в день (SleepTime) на основе других факторов." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Формируем выборки" + ] + }, + { + "cell_type": "code", + "execution_count": 85, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'X_train'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
StateSexGeneralHealthPhysicalHealthDaysMentalHealthDaysLastCheckupTimePhysicalActivitiesRemovedTeethHadHeartAttackHadAngina...HeightInMetersWeightInKilogramsBMIAlcoholDrinkersHIVTestingFluVaxLast12PneumoVaxEverTetanusLast10TdapHighRiskLastYearCovidPos
108769MinnesotaMaleGood0.00.0Within past year (anytime less than 12 months ...YesNone of themNoNo...1.7383.9128.13NoYesYesYesYes, received tetanus shot but not sure what typeYesYes
240750GuamMaleExcellent0.00.0Within past 2 years (1 year but less than 2 ye...YesNone of themNoYes...1.6570.3125.79YesNoNoNoNo, did not receive any tetanus shot in the pa...NoYes
100329MichiganFemaleExcellent3.00.0Within past year (anytime less than 12 months ...No1 to 5NoYes...1.6058.9723.03NoNoYesYesYes, received tetanus shot but not sure what typeNoNo
132628New HampshireMaleGood4.06.0Within past year (anytime less than 12 months ...Yes1 to 5NoYes...1.7068.0423.49YesNoYesNoNo, did not receive any tetanus shot in the pa...NoNo
72101KansasMaleVery good0.02.0Within past year (anytime less than 12 months ...YesNone of themNoNo...1.8399.7929.84YesNoYesYesYes, received TdapNoNo
..................................................................
119879MissouriFemaleExcellent0.00.0Within past year (anytime less than 12 months ...YesNone of themNoNo...1.7861.2319.37YesYesYesNoYes, received tetanus shot but not sure what typeNoNo
103694MichiganFemaleGood10.00.0Within past year (anytime less than 12 months ...Yes1 to 5NoNo...1.6374.8428.32YesNoYesYesNo, did not receive any tetanus shot in the pa...NoNo
131932NevadaFemaleGood0.00.0Within past year (anytime less than 12 months ...Yes1 to 5NoNo...1.7090.7231.32NoNoNoNoNo, did not receive any tetanus shot in the pa...NoNo
146867New YorkFemaleVery good0.00.0Within past year (anytime less than 12 months ...Yes1 to 5NoNo...1.6877.1127.44YesNoYesNoNo, did not receive any tetanus shot in the pa...NoYes
121958MontanaFemaleGood1.00.0Within past year (anytime less than 12 months ...YesAllNoNo...1.6598.8836.28YesYesYesYesYes, received tetanus shot but not sure what typeNoNo
\n", + "

196817 rows × 39 columns

\n", + "
" + ], + "text/plain": [ + " State Sex GeneralHealth PhysicalHealthDays \\\n", + "108769 Minnesota Male Good 0.0 \n", + "240750 Guam Male Excellent 0.0 \n", + "100329 Michigan Female Excellent 3.0 \n", + "132628 New Hampshire Male Good 4.0 \n", + "72101 Kansas Male Very good 0.0 \n", + "... ... ... ... ... \n", + "119879 Missouri Female Excellent 0.0 \n", + "103694 Michigan Female Good 10.0 \n", + "131932 Nevada Female Good 0.0 \n", + "146867 New York Female Very good 0.0 \n", + "121958 Montana Female Good 1.0 \n", + "\n", + " MentalHealthDays LastCheckupTime \\\n", + "108769 0.0 Within past year (anytime less than 12 months ... \n", + "240750 0.0 Within past 2 years (1 year but less than 2 ye... \n", + "100329 0.0 Within past year (anytime less than 12 months ... \n", + "132628 6.0 Within past year (anytime less than 12 months ... \n", + "72101 2.0 Within past year (anytime less than 12 months ... \n", + "... ... ... \n", + "119879 0.0 Within past year (anytime less than 12 months ... \n", + "103694 0.0 Within past year (anytime less than 12 months ... \n", + "131932 0.0 Within past year (anytime less than 12 months ... \n", + "146867 0.0 Within past year (anytime less than 12 months ... \n", + "121958 0.0 Within past year (anytime less than 12 months ... \n", + "\n", + " PhysicalActivities RemovedTeeth HadHeartAttack HadAngina ... \\\n", + "108769 Yes None of them No No ... \n", + "240750 Yes None of them No Yes ... \n", + "100329 No 1 to 5 No Yes ... \n", + "132628 Yes 1 to 5 No Yes ... \n", + "72101 Yes None of them No No ... \n", + "... ... ... ... ... ... \n", + "119879 Yes None of them No No ... \n", + "103694 Yes 1 to 5 No No ... \n", + "131932 Yes 1 to 5 No No ... \n", + "146867 Yes 1 to 5 No No ... \n", + "121958 Yes All No No ... \n", + "\n", + " HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting \\\n", + "108769 1.73 83.91 28.13 No Yes \n", + "240750 1.65 70.31 25.79 Yes No \n", + "100329 1.60 58.97 23.03 No No \n", + "132628 1.70 68.04 23.49 Yes No \n", + "72101 1.83 99.79 29.84 Yes No \n", + "... ... ... ... ... ... \n", + "119879 1.78 61.23 19.37 Yes Yes \n", + "103694 1.63 74.84 28.32 Yes No \n", + "131932 1.70 90.72 31.32 No No \n", + "146867 1.68 77.11 27.44 Yes No \n", + "121958 1.65 98.88 36.28 Yes Yes \n", + "\n", + " FluVaxLast12 PneumoVaxEver \\\n", + "108769 Yes Yes \n", + "240750 No No \n", + "100329 Yes Yes \n", + "132628 Yes No \n", + "72101 Yes Yes \n", + "... ... ... \n", + "119879 Yes No \n", + "103694 Yes Yes \n", + "131932 No No \n", + "146867 Yes No \n", + "121958 Yes Yes \n", + "\n", + " TetanusLast10Tdap HighRiskLastYear \\\n", + "108769 Yes, received tetanus shot but not sure what type Yes \n", + "240750 No, did not receive any tetanus shot in the pa... No \n", + "100329 Yes, received tetanus shot but not sure what type No \n", + "132628 No, did not receive any tetanus shot in the pa... No \n", + "72101 Yes, received Tdap No \n", + "... ... ... \n", + "119879 Yes, received tetanus shot but not sure what type No \n", + "103694 No, did not receive any tetanus shot in the pa... No \n", + "131932 No, did not receive any tetanus shot in the pa... No \n", + "146867 No, did not receive any tetanus shot in the pa... No \n", + "121958 Yes, received tetanus shot but not sure what type No \n", + "\n", + " CovidPos \n", + "108769 Yes \n", + "240750 Yes \n", + "100329 No \n", + "132628 No \n", + "72101 No \n", + "... ... \n", + "119879 No \n", + "103694 No \n", + "131932 No \n", + "146867 Yes \n", + "121958 No \n", + "\n", + "[196817 rows x 39 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'y_train'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
SleepHours
1087696.0
2407507.0
1003299.0
1326286.0
721017.0
......
1198798.0
1036948.0
1319327.0
1468678.0
1219588.0
\n", + "

196817 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " SleepHours\n", + "108769 6.0\n", + "240750 7.0\n", + "100329 9.0\n", + "132628 6.0\n", + "72101 7.0\n", + "... ...\n", + "119879 8.0\n", + "103694 8.0\n", + "131932 7.0\n", + "146867 8.0\n", + "121958 8.0\n", + "\n", + "[196817 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'X_test'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
StateSexGeneralHealthPhysicalHealthDaysMentalHealthDaysLastCheckupTimePhysicalActivitiesRemovedTeethHadHeartAttackHadAngina...HeightInMetersWeightInKilogramsBMIAlcoholDrinkersHIVTestingFluVaxLast12PneumoVaxEverTetanusLast10TdapHighRiskLastYearCovidPos
194767TexasFemaleGood0.00.0Within past year (anytime less than 12 months ...YesNone of themNoNo...1.68113.4040.35NoNoNoNoNo, did not receive any tetanus shot in the pa...NoYes
231923WisconsinFemaleGood2.05.0Within past year (anytime less than 12 months ...Yes1 to 5NoNo...1.73104.3334.97YesYesNoYesNo, did not receive any tetanus shot in the pa...NoNo
52815IdahoMalePoor7.010.0Within past year (anytime less than 12 months ...Yes1 to 5NoYes...1.73104.3334.97NoNoYesYesYes, received tetanus shot but not sure what typeNoNo
65909IowaFemaleGood20.010.0Within past year (anytime less than 12 months ...NoAllYesNo...1.68127.0145.19NoNoNoNoNo, did not receive any tetanus shot in the pa...NoYes
184154South DakotaFemaleExcellent0.00.0Within past year (anytime less than 12 months ...YesNone of themNoNo...1.6049.9019.49YesNoYesNoYes, received TdapNoTested positive using home test without a heal...
..................................................................
57503IndianaFemaleFair3.00.0Within past year (anytime less than 12 months ...Yes1 to 5YesNo...1.6397.5236.90YesYesYesYesYes, received TdapNoNo
47420HawaiiFemaleFair30.05.0Within past year (anytime less than 12 months ...YesNone of themNoNo...1.7077.5626.78NoYesYesNoYes, received tetanus shot but not sure what typeNoNo
186088South DakotaFemaleGood15.015.0Within past year (anytime less than 12 months ...Yes1 to 5NoNo...1.7354.8818.40YesNoYesNoYes, received tetanus shot but not sure what typeNoYes
11687ArkansasMaleExcellent0.00.0Within past year (anytime less than 12 months ...YesNone of themNoNo...1.7888.4527.98YesYesYesNoYes, received tetanus shot but not sure what typeNoNo
200835UtahMaleVery good0.00.0Within past year (anytime less than 12 months ...YesNone of themNoNo...1.91118.3932.62NoYesNoYesYes, received TdapNoYes
\n", + "

49205 rows × 39 columns

\n", + "
" + ], + "text/plain": [ + " State Sex GeneralHealth PhysicalHealthDays \\\n", + "194767 Texas Female Good 0.0 \n", + "231923 Wisconsin Female Good 2.0 \n", + "52815 Idaho Male Poor 7.0 \n", + "65909 Iowa Female Good 20.0 \n", + "184154 South Dakota Female Excellent 0.0 \n", + "... ... ... ... ... \n", + "57503 Indiana Female Fair 3.0 \n", + "47420 Hawaii Female Fair 30.0 \n", + "186088 South Dakota Female Good 15.0 \n", + "11687 Arkansas Male Excellent 0.0 \n", + "200835 Utah Male Very good 0.0 \n", + "\n", + " MentalHealthDays LastCheckupTime \\\n", + "194767 0.0 Within past year (anytime less than 12 months ... \n", + "231923 5.0 Within past year (anytime less than 12 months ... \n", + "52815 10.0 Within past year (anytime less than 12 months ... \n", + "65909 10.0 Within past year (anytime less than 12 months ... \n", + "184154 0.0 Within past year (anytime less than 12 months ... \n", + "... ... ... \n", + "57503 0.0 Within past year (anytime less than 12 months ... \n", + "47420 5.0 Within past year (anytime less than 12 months ... \n", + "186088 15.0 Within past year (anytime less than 12 months ... \n", + "11687 0.0 Within past year (anytime less than 12 months ... \n", + "200835 0.0 Within past year (anytime less than 12 months ... \n", + "\n", + " PhysicalActivities RemovedTeeth HadHeartAttack HadAngina ... \\\n", + "194767 Yes None of them No No ... \n", + "231923 Yes 1 to 5 No No ... \n", + "52815 Yes 1 to 5 No Yes ... \n", + "65909 No All Yes No ... \n", + "184154 Yes None of them No No ... \n", + "... ... ... ... ... ... \n", + "57503 Yes 1 to 5 Yes No ... \n", + "47420 Yes None of them No No ... \n", + "186088 Yes 1 to 5 No No ... \n", + "11687 Yes None of them No No ... \n", + "200835 Yes None of them No No ... \n", + "\n", + " HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting \\\n", + "194767 1.68 113.40 40.35 No No \n", + "231923 1.73 104.33 34.97 Yes Yes \n", + "52815 1.73 104.33 34.97 No No \n", + "65909 1.68 127.01 45.19 No No \n", + "184154 1.60 49.90 19.49 Yes No \n", + "... ... ... ... ... ... \n", + "57503 1.63 97.52 36.90 Yes Yes \n", + "47420 1.70 77.56 26.78 No Yes \n", + "186088 1.73 54.88 18.40 Yes No \n", + "11687 1.78 88.45 27.98 Yes Yes \n", + "200835 1.91 118.39 32.62 No Yes \n", + "\n", + " FluVaxLast12 PneumoVaxEver \\\n", + "194767 No No \n", + "231923 No Yes \n", + "52815 Yes Yes \n", + "65909 No No \n", + "184154 Yes No \n", + "... ... ... \n", + "57503 Yes Yes \n", + "47420 Yes No \n", + "186088 Yes No \n", + "11687 Yes No \n", + "200835 No Yes \n", + "\n", + " TetanusLast10Tdap HighRiskLastYear \\\n", + "194767 No, did not receive any tetanus shot in the pa... No \n", + "231923 No, did not receive any tetanus shot in the pa... No \n", + "52815 Yes, received tetanus shot but not sure what type No \n", + "65909 No, did not receive any tetanus shot in the pa... No \n", + "184154 Yes, received Tdap No \n", + "... ... ... \n", + "57503 Yes, received Tdap No \n", + "47420 Yes, received tetanus shot but not sure what type No \n", + "186088 Yes, received tetanus shot but not sure what type No \n", + "11687 Yes, received tetanus shot but not sure what type No \n", + "200835 Yes, received Tdap No \n", + "\n", + " CovidPos \n", + "194767 Yes \n", + "231923 No \n", + "52815 No \n", + "65909 Yes \n", + "184154 Tested positive using home test without a heal... \n", + "... ... \n", + "57503 No \n", + "47420 No \n", + "186088 Yes \n", + "11687 No \n", + "200835 Yes \n", + "\n", + "[49205 rows x 39 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'y_test'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
SleepHours
1947678.0
2319238.0
528156.0
659098.0
1841547.0
......
575036.0
474206.0
1860886.0
116878.0
2008358.0
\n", + "

49205 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " SleepHours\n", + "194767 8.0\n", + "231923 8.0\n", + "52815 6.0\n", + "65909 8.0\n", + "184154 7.0\n", + "... ...\n", + "57503 6.0\n", + "47420 6.0\n", + "186088 6.0\n", + "11687 8.0\n", + "200835 8.0\n", + "\n", + "[49205 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "df = pd.read_csv(\"csv\\\\heart_2022_no_nans.csv\")\n", + "\n", + "TARGET_COLUMN_NAME_REGRESSION = \"SleepHours\"\n", + "\n", + "def split_into_train_test(\n", + " df_input: DataFrame,\n", + " target_colname: str,\n", + " frac_train: float = 0.8,\n", + " random_state: int = None,\n", + ") -> Tuple[DataFrame, DataFrame, DataFrame, DataFrame]:\n", + " \n", + " if not (0 < frac_train < 1):\n", + " raise ValueError(\"Fraction must be between 0 and 1.\")\n", + " \n", + " if target_colname not in df_input.columns:\n", + " raise ValueError(f\"{target_colname} is not a column in the DataFrame.\")\n", + " \n", + " X = df_input.drop(columns=[target_colname])\n", + " y = df_input[[target_colname]]\n", + "\n", + " X_train, X_test, y_train, y_test = train_test_split(\n", + " X, y,\n", + " test_size=(1.0 - frac_train),\n", + " random_state=random_state\n", + " )\n", + " return X_train, X_test, y_train, y_test\n", + "\n", + "X_train, X_test, y_train, y_test = split_into_train_test(\n", + " df, \n", + " target_colname=TARGET_COLUMN_NAME_REGRESSION, \n", + " frac_train=0.8, \n", + " random_state=42\n", + ")\n", + "\n", + "display(\"X_train\", X_train)\n", + "display(\"y_train\", y_train)\n", + "\n", + "display(\"X_test\", X_test)\n", + "display(\"y_test\", y_test)" + ] + }, + { + "cell_type": "code", + "execution_count": 86, + "metadata": {}, + "outputs": [], + "source": [ + "def get_filtered_columns(df: DataFrame, no_numeric=False, no_text=False) -> list[str]:\n", + " \"\"\"\n", + " Возвращает список колонок по фильтру\n", + " \"\"\"\n", + " w = []\n", + " for column in df.columns:\n", + " if no_numeric and pd.api.types.is_numeric_dtype(df[column]):\n", + " continue\n", + " if no_text and not pd.api.types.is_numeric_dtype(df[column]):\n", + " continue\n", + " w.append(column)\n", + " return w" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Выполним one-hot encoding, чтобы избавиться от категориальных признаков" + ] + }, + { + "cell_type": "code", + "execution_count": 87, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
PhysicalHealthDaysMentalHealthDaysHeightInMetersWeightInKilogramsBMIState_AlaskaState_ArizonaState_ArkansasState_CaliforniaState_Colorado...AlcoholDrinkers_YesHIVTesting_YesFluVaxLast12_YesPneumoVaxEver_YesTetanusLast10Tdap_Yes, received TdapTetanusLast10Tdap_Yes, received tetanus shot but not sure what typeTetanusLast10Tdap_Yes, received tetanus shot, but not TdapHighRiskLastYear_YesCovidPos_Tested positive using home test without a health professionalCovidPos_Yes
1087690.00.01.7383.9128.13FalseFalseFalseFalseFalse...FalseTrueTrueTrueFalseTrueFalseTrueFalseTrue
2407500.00.01.6570.3125.79FalseFalseFalseFalseFalse...TrueFalseFalseFalseFalseFalseFalseFalseFalseTrue
1003293.00.01.6058.9723.03FalseFalseFalseFalseFalse...FalseFalseTrueTrueFalseTrueFalseFalseFalseFalse
1326284.06.01.7068.0423.49FalseFalseFalseFalseFalse...TrueFalseTrueFalseFalseFalseFalseFalseFalseFalse
721010.02.01.8399.7929.84FalseFalseFalseFalseFalse...TrueFalseTrueTrueTrueFalseFalseFalseFalseFalse
..................................................................
1198790.00.01.7861.2319.37FalseFalseFalseFalseFalse...TrueTrueTrueFalseFalseTrueFalseFalseFalseFalse
10369410.00.01.6374.8428.32FalseFalseFalseFalseFalse...TrueFalseTrueTrueFalseFalseFalseFalseFalseFalse
1319320.00.01.7090.7231.32FalseFalseFalseFalseFalse...FalseFalseFalseFalseFalseFalseFalseFalseFalseFalse
1468670.00.01.6877.1127.44FalseFalseFalseFalseFalse...TrueFalseTrueFalseFalseFalseFalseFalseFalseTrue
1219581.00.01.6598.8836.28FalseFalseFalseFalseFalse...TrueTrueTrueTrueFalseTrueFalseFalseFalseFalse
\n", + "

196817 rows × 121 columns

\n", + "
" + ], + "text/plain": [ + " PhysicalHealthDays MentalHealthDays HeightInMeters \\\n", + "108769 0.0 0.0 1.73 \n", + "240750 0.0 0.0 1.65 \n", + "100329 3.0 0.0 1.60 \n", + "132628 4.0 6.0 1.70 \n", + "72101 0.0 2.0 1.83 \n", + "... ... ... ... \n", + "119879 0.0 0.0 1.78 \n", + "103694 10.0 0.0 1.63 \n", + "131932 0.0 0.0 1.70 \n", + "146867 0.0 0.0 1.68 \n", + "121958 1.0 0.0 1.65 \n", + "\n", + " WeightInKilograms BMI State_Alaska State_Arizona State_Arkansas \\\n", + "108769 83.91 28.13 False False False \n", + "240750 70.31 25.79 False False False \n", + "100329 58.97 23.03 False False False \n", + "132628 68.04 23.49 False False False \n", + "72101 99.79 29.84 False False False \n", + "... ... ... ... ... ... \n", + "119879 61.23 19.37 False False False \n", + "103694 74.84 28.32 False False False \n", + "131932 90.72 31.32 False False False \n", + "146867 77.11 27.44 False False False \n", + "121958 98.88 36.28 False False False \n", + "\n", + " State_California State_Colorado ... AlcoholDrinkers_Yes \\\n", + "108769 False False ... False \n", + "240750 False False ... True \n", + "100329 False False ... False \n", + "132628 False False ... True \n", + "72101 False False ... True \n", + "... ... ... ... ... \n", + "119879 False False ... True \n", + "103694 False False ... True \n", + "131932 False False ... False \n", + "146867 False False ... True \n", + "121958 False False ... True \n", + "\n", + " HIVTesting_Yes FluVaxLast12_Yes PneumoVaxEver_Yes \\\n", + "108769 True True True \n", + "240750 False False False \n", + "100329 False True True \n", + "132628 False True False \n", + "72101 False True True \n", + "... ... ... ... \n", + "119879 True True False \n", + "103694 False True True \n", + "131932 False False False \n", + "146867 False True False \n", + "121958 True True True \n", + "\n", + " TetanusLast10Tdap_Yes, received Tdap \\\n", + "108769 False \n", + "240750 False \n", + "100329 False \n", + "132628 False \n", + "72101 True \n", + "... ... \n", + "119879 False \n", + "103694 False \n", + "131932 False \n", + "146867 False \n", + "121958 False \n", + "\n", + " TetanusLast10Tdap_Yes, received tetanus shot but not sure what type \\\n", + "108769 True \n", + "240750 False \n", + "100329 True \n", + "132628 False \n", + "72101 False \n", + "... ... \n", + "119879 True \n", + "103694 False \n", + "131932 False \n", + "146867 False \n", + "121958 True \n", + "\n", + " TetanusLast10Tdap_Yes, received tetanus shot, but not Tdap \\\n", + "108769 False \n", + "240750 False \n", + "100329 False \n", + "132628 False \n", + "72101 False \n", + "... ... \n", + "119879 False \n", + "103694 False \n", + "131932 False \n", + "146867 False \n", + "121958 False \n", + "\n", + " HighRiskLastYear_Yes \\\n", + "108769 True \n", + "240750 False \n", + "100329 False \n", + "132628 False \n", + "72101 False \n", + "... ... \n", + "119879 False \n", + "103694 False \n", + "131932 False \n", + "146867 False \n", + "121958 False \n", + "\n", + " CovidPos_Tested positive using home test without a health professional \\\n", + "108769 False \n", + "240750 False \n", + "100329 False \n", + "132628 False \n", + "72101 False \n", + "... ... \n", + "119879 False \n", + "103694 False \n", + "131932 False \n", + "146867 False \n", + "121958 False \n", + "\n", + " CovidPos_Yes \n", + "108769 True \n", + "240750 True \n", + "100329 False \n", + "132628 False \n", + "72101 False \n", + "... ... \n", + "119879 False \n", + "103694 False \n", + "131932 False \n", + "146867 True \n", + "121958 False \n", + "\n", + "[196817 rows x 121 columns]" + ] + }, + "execution_count": 87, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cat_features = get_filtered_columns(df, no_numeric=True)\n", + "\n", + "X_test = pd.get_dummies(X_test, columns=cat_features, drop_first=True)\n", + "X_train = pd.get_dummies(X_train, columns=cat_features, drop_first=True)\n", + "\n", + "X_test\n", + "X_train" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Определение перечня алгоритмов решения задачи регрессии" + ] + }, + { + "cell_type": "code", + "execution_count": 88, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Model: linear\n", + "Model: linear_poly\n", + "Model: linear_interact\n", + "Model: ridge\n", + "Model: decision_tree\n", + "Model: knn\n", + "Model: random_forest\n", + "Model: mlp\n" + ] + } + ], + "source": [ + "import math\n", + "from sklearn.pipeline import make_pipeline\n", + "from sklearn.preprocessing import PolynomialFeatures\n", + "\n", + "\n", + "models = {\n", + " \"linear\": {\"model\": linear_model.LinearRegression(n_jobs=-1)},\n", + " \"linear_poly\": {\n", + " \"model\": make_pipeline(\n", + " PolynomialFeatures(degree=2),\n", + " linear_model.LinearRegression(fit_intercept=False, n_jobs=-1),\n", + " )\n", + " },\n", + " \"linear_interact\": {\n", + " \"model\": make_pipeline(\n", + " PolynomialFeatures(interaction_only=True),\n", + " linear_model.LinearRegression(fit_intercept=False, n_jobs=-1),\n", + " )\n", + " },\n", + " \"ridge\": {\"model\": linear_model.RidgeCV()},\n", + " \"decision_tree\": {\n", + " \"model\": tree.DecisionTreeRegressor(max_depth=7, random_state=random_state)\n", + " },\n", + " \"knn\": {\"model\": neighbors.KNeighborsRegressor(n_neighbors=7, n_jobs=-1)},\n", + " \"random_forest\": {\n", + " \"model\": ensemble.RandomForestRegressor(\n", + " max_depth=7, random_state=random_state, n_jobs=-1\n", + " )\n", + " },\n", + " \"mlp\": {\n", + " \"model\": neural_network.MLPRegressor(\n", + " activation=\"tanh\",\n", + " hidden_layer_sizes=(3),\n", + " max_iter=500,\n", + " early_stopping=True,\n", + " random_state=random_state,\n", + " )\n", + " },\n", + "}\n", + "\n", + "for model_name in models.keys():\n", + " print(f\"Model: {model_name}\")\n", + "\n", + " fitted_model = models[model_name][\"model\"].fit(\n", + " X_train.values, y_train.values.ravel()\n", + " )\n", + " y_train_pred = fitted_model.predict(X_train.values)\n", + " y_test_pred = fitted_model.predict(X_test.values)\n", + " models[model_name][\"fitted\"] = fitted_model\n", + " models[model_name][\"train_preds\"] = y_train_pred\n", + " models[model_name][\"preds\"] = y_test_pred\n", + " models[model_name][\"RMSE_train\"] = math.sqrt(\n", + " metrics.mean_squared_error(y_train, y_train_pred)\n", + " )\n", + " models[model_name][\"RMSE_test\"] = math.sqrt(\n", + " metrics.mean_squared_error(y_test, y_test_pred)\n", + " )\n", + " models[model_name][\"RMAE_test\"] = math.sqrt(\n", + " metrics.mean_absolute_error(y_test, y_test_pred)\n", + " )\n", + " models[model_name][\"R2_test\"] = metrics.r2_score(y_test, y_test_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Выводим результаты оценки" + ] + }, + { + "cell_type": "code", + "execution_count": 89, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
 RMSE_trainRMSE_testRMAE_testR2_test
mlp1.4015711.4015561.0018320.049273
ridge1.4031851.4018591.0018850.048861
linear1.4031841.4018601.0018980.048860
random_forest1.4003601.4081851.0014820.040258
linear_poly1.3659121.4166531.0083700.028680
linear_interact1.3660661.4170081.0085430.028193
decision_tree1.4060261.4177501.0055760.027175
knn1.2964921.4933161.041495-0.079292
\n" + ], + "text/plain": [ + "" + ] + }, + "execution_count": 89, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "reg_metrics = pd.DataFrame.from_dict(models, \"index\")[\n", + " [\"RMSE_train\", \"RMSE_test\", \"RMAE_test\", \"R2_test\"]\n", + "]\n", + "reg_metrics.sort_values(by=\"RMSE_test\").style.background_gradient(\n", + " cmap=\"viridis\", low=1, high=0.3, subset=[\"RMSE_train\", \"RMSE_test\"]\n", + ").background_gradient(cmap=\"plasma\", low=0.3, high=1, subset=[\"RMAE_test\", \"R2_test\"])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Выводим лучшую модель" + ] + }, + { + "cell_type": "code", + "execution_count": 90, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'mlp'" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "best_model = str(reg_metrics.sort_values(by=\"RMSE_test\").iloc[0].name)\n", + "\n", + "display(best_model)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Подбираем гиперпараметры методом поиска по сетке" + ] + }, + { + "cell_type": "code", + "execution_count": 92, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Fitting 3 folds for each of 8 candidates, totalling 24 fits\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "d:\\code\\AIM-PIbd-31-Potapov-N-S\\lab_4\\.venv\\Lib\\site-packages\\sklearn\\base.py:1473: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n", + " return fit_method(estimator, *args, **kwargs)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Лучшие параметры: {'max_depth': 10, 'min_samples_split': 5, 'n_estimators': 100}\n", + "Лучший результат (MSE): 1.9866610870680514\n" + ] + } + ], + "source": [ + "from sklearn.ensemble import RandomForestRegressor\n", + "\n", + "\n", + "X = df[get_filtered_columns(df, no_numeric=True)]\n", + "y = df[TARGET_COLUMN_NAME_REGRESSION] \n", + "\n", + "model = RandomForestRegressor() \n", + "\n", + "param_grid = {\n", + " 'n_estimators': [50, 100], \n", + " 'max_depth': [10, 20], \n", + " 'min_samples_split': [5, 10] \n", + "}\n", + "\n", + "grid_search = GridSearchCV(estimator=model, param_grid=param_grid,\n", + " scoring='neg_mean_squared_error', cv=3, n_jobs=-1, verbose=2)\n", + "\n", + "grid_search.fit(X_train, y_train)\n", + "\n", + "print(\"Лучшие параметры:\", grid_search.best_params_)\n", + "print(\"Лучший результат (MSE):\", -grid_search.best_score_)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Обучаем модель с новыми гиперпараметрами и сравниваем новых данных со старыми" + ] + }, + { + "cell_type": "code", + "execution_count": 93, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Fitting 3 folds for each of 8 candidates, totalling 24 fits\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "d:\\code\\AIM-PIbd-31-Potapov-N-S\\lab_4\\.venv\\Lib\\site-packages\\sklearn\\base.py:1473: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n", + " return fit_method(estimator, *args, **kwargs)\n", + "d:\\code\\AIM-PIbd-31-Potapov-N-S\\lab_4\\.venv\\Lib\\site-packages\\sklearn\\base.py:1473: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n", + " return fit_method(estimator, *args, **kwargs)\n", + "d:\\code\\AIM-PIbd-31-Potapov-N-S\\lab_4\\.venv\\Lib\\site-packages\\sklearn\\base.py:1473: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n", + " return fit_method(estimator, *args, **kwargs)\n", + "d:\\code\\AIM-PIbd-31-Potapov-N-S\\lab_4\\.venv\\Lib\\site-packages\\sklearn\\base.py:1473: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n", + " return fit_method(estimator, *args, **kwargs)\n", + "d:\\code\\AIM-PIbd-31-Potapov-N-S\\lab_4\\.venv\\Lib\\site-packages\\sklearn\\base.py:1473: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n", + " return fit_method(estimator, *args, **kwargs)\n", + "d:\\code\\AIM-PIbd-31-Potapov-N-S\\lab_4\\.venv\\Lib\\site-packages\\sklearn\\base.py:1473: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n", + " return fit_method(estimator, *args, **kwargs)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Старые параметры: {'max_depth': 10, 'min_samples_split': 10, 'n_estimators': 100}\n", + "Лучший результат (MSE) на старых параметрах: 1.9867639342405718\n", + "\n", + "Новые параметры: {'max_depth': 10, 'min_samples_split': 5, 'n_estimators': 100}\n", + "Лучший результат (MSE) на новых параметрах: 1.990467882679972\n", + "Среднеквадратическая ошибка (MSE) на тестовых данных: 1.975249119855746\n", + "Корень среднеквадратичной ошибки (RMSE) на тестовых данных: 1.4054355623278307\n" + ] + } + ], + "source": [ + "# Old data\n", + "\n", + "old_param_grid = param_grid\n", + "old_grid_search = grid_search\n", + "old_grid_search.fit(X_train, y_train)\n", + "\n", + "old_best_params = old_grid_search.best_params_\n", + "old_best_mse = -old_grid_search.best_score_ \n", + "\n", + "# New data\n", + "\n", + "new_param_grid = {\n", + " 'n_estimators': [100],\n", + " 'max_depth': [10],\n", + " 'min_samples_split': [5]\n", + " }\n", + "new_grid_search = GridSearchCV(estimator=RandomForestRegressor(), \n", + " param_grid=new_param_grid,\n", + " scoring='neg_mean_squared_error', cv=2)\n", + "\n", + "new_grid_search.fit(X_train, y_train)\n", + "\n", + "new_best_params = new_grid_search.best_params_\n", + "new_best_mse = -new_grid_search.best_score_\n", + "\n", + "new_best_model = RandomForestRegressor(**new_best_params)\n", + "new_best_model.fit(X_train, y_train)\n", + "\n", + "old_best_model = RandomForestRegressor(**old_best_params)\n", + "old_best_model.fit(X_train, y_train)\n", + "\n", + "y_new_pred = new_best_model.predict(X_test)\n", + "y_old_pred = old_best_model.predict(X_test)\n", + "\n", + "mse = metrics.mean_squared_error(y_test, y_new_pred)\n", + "rmse = np.sqrt(mse)\n", + "\n", + "print(\"Старые параметры:\", old_best_params)\n", + "print(\"Лучший результат (MSE) на старых параметрах:\", old_best_mse)\n", + "print(\"\\nНовые параметры:\", new_best_params)\n", + "print(\"Лучший результат (MSE) на новых параметрах:\", new_best_mse)\n", + "print(\"Среднеквадратическая ошибка (MSE) на тестовых данных:\", mse)\n", + "print(\"Корень среднеквадратичной ошибки (RMSE) на тестовых данных:\", rmse)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Визуализация данных" + ] + }, + { + "cell_type": "code", + "execution_count": 94, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "plt.figure(figsize=(16, 8))\n", + "plt.scatter(range(len(y_test)), y_test, label=\"Истинные значения\", color=\"black\", alpha=0.5)\n", + "plt.scatter(range(len(y_test)), y_new_pred, label=\"Предсказанные (новые параметры)\", color=\"blue\", alpha=0.5)\n", + "plt.scatter(range(len(y_test)), y_old_pred, label=\"Предсказанные (старые параметры)\", color=\"red\", alpha=0.5)\n", + "plt.xlabel(\"Выборка\")\n", + "plt.ylabel(\"Значения\")\n", + "plt.legend()\n", + "plt.title(\"Сравнение предсказанных и истинных значений\")\n", + "plt.show()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": ".venv", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.6" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/lab_4/requirements.txt b/lab_4/requirements.txt new file mode 100644 index 0000000000000000000000000000000000000000..692ac49df2348a47a070471305b6aa9aed8ed7fa GIT binary patch literal 2706 zcmZ{m-EPxR5QNV)5|0wpq!&ZYRaJu`q`C-vMQba_g+7X@}^wqnU~k)RcRt(Kb>(x#95r)>RwZx6tgL7eQp)E z(`T;?S3&&T*}+-S>}IKPTInXT1*ZZCCzQ(B=Snmp&v;&u_aSM5hVM4TytYFAOG zRh~!vw1b>>4aQnxKl;&&<&Z4cC9Mtcpu!OBJ8dvX52gc zWIUP;%DGn*_+Zo8n}^ffJNd#MsZsEJIckDO?xNFPRliBMpHm-;P{qoBi7FfQ+K63O z1Jl*W$3}Q>b+*;hsRLM*;*4JAaohTSP?b)aG`hjvbg+@%E9dQq znNt`CpFR3NB^^-0!sgCj3NpyQJKQIkLuyUh&Fv?aK z0zbXOkk$M*8kc)k@i`o2FQ<7?7nP`VC-t1ax}o;=pz23GXt|ZIau$=R>u|tF<=bmd zP~)?HUym4aLOhS`h(Qr-mvl;Mg5E~@Qq)Wcx?lpIaK}2+A*?+HLS`w?-9hH;)a`R7 zoq>PtBFCVQyENb40roAGh*}H(UNz|kEZ!(&va7(rCYUMWAcmPEluw6|Eo(l-?0EYn zTXt@*+w9&(3ekP;9NbqUx|rm26#b?*xXo*W>AH?FFYHr@N@K(yQwa5mZQ+&lBF{E`B}PS`2VKCb(`PWmzUSSlYo>>=>jF?4NC{s8Ykk0Jm7 literal 0 HcmV?d00001